1. Field of the Invention
The present invention relates generally to a data recording method, data reproduction method, data recording device, data reproduction device and optical recording medium, and more particularly, to a data recording method, data reproduction method, data recording device, data reproduction device and optical recording medium adapted to high-density data recording.
2. Description of Related Art
As one type of recording method for recording data onto an optical recording medium (hereinafter optical disk), a so-called mark edge recording system is known. In the mark edge recording system, the edge portions of the marks recorded on the optical disk represent the data value. Standards for recording data on an optical disk using the mark edge recording system have been proposed (see “Data Interchange on 90 mm Optical Disk Cartridges”, ISO/IEC JTC 1/SUBTRACTING CIRCUIT 23 N705, 1.23.06 Draft 2 December 1994.)
However, in the mark edge recording system, when an RLL (Run Length Limited) (1,7) modulation code is recorded and the number of data logic value ones (“1”s) (hereinafter positive data) and the number of data logic value zeros (“0”s) (hereinafter negative data) differ drastically within a given sector, the average value of the direct current component of a reproduction signal from the optical disk (hereinafter DC component) is either too large or too small. In such cases, properly setting the slice level used to digitize the reproduction signal is difficult, making accurate reproduction of the data problematic.
The above-mentioned standards deal with this problem by switching as appropriate a resync pattern inserted between the data blocks, so that the accumulated sum of the DC component of the reproduction signal approaches a constant value (for example “0”) and the slice level margin at time of data reproduction is increased. Specific means by which this action is accomplished is taught for example by Japanese Laid-Open Patent Application No. 8-279251.
Additionally, as a precision recording system for recording and reproducing data, the so-called PRML (partial response maximum likelihood) system is known. In the PRML system, the recording data is modulated into a partial response wave pattern and recorded onto a magneto-optical disk. The reproduction signals from that magneto-optical disk are sampled at predetermined intervals, after which data of maximum likelihood are detected by a Viterbi detector.
In the PRML system, the reproduction signal is sampled at predetermined intervals and data reproduction is carried out by detecting the transition in the level of the sampled values, so changes in the DC component of the reproduction signal degrade the potential to reproduce the data accurately. Hence the PRML system also inserts a resync pattern between the data blocks so as to minimize the DC component within a given sector when recording data to the magneto-optical disk.
Conventionally, in the PRML system, by switching the resync pattern inserted between the data blocks as appropriate, the accumulated sum of the DC component of the reproduction signal approaches a constant value (for example “0”) and thus the DC component within any given sector can be held to a minimum.
Additionally, in the PRML system, when reproducing data from a magneto-optical disk, the amount by which the DC component of the reproduction wave form changes (hereinafter referred to as the offset amount) is calculated using the moving average method. By feeding the offset amount so calculated back to the expected value of the reproduction system (that is, the Viterbi decoder) the impact of the offset amount on the reproduction wave form is reduced.
However, the above-described method of minimizing the DC component within a sector by switching the resync pattern inserted between data blocks as appropriate has a disadvantage in that the DC component can change dramatically between data blocks. With the PRML system, this type of drastic fluctuation in DC component between data blocks degrades the accuracy with which data can be reproduced.
Additionally, with a Viterbi detection circuit, in which the expected value can be changed, though it is possible to calculate the sector-specific DC fluctuation component (that is, the offset) and dynamically reflect that in the expected value, the moving average method of calculating the offset amount cannot be used if the DC component fluctuates drastically between data blocks.
In particular, the method of appropriately switching the resync pattern to be inserted between the data blocks so as to minimize the DC component within a given sector sometimes causes sharp fluctuations in the DC component before and after the resync pattern located between data blocks. Such sharp fluctuations cannot be accommodated using the moving average method of calculating the offset amount.